Skip to main content

Foundational Trials

DanGer Shock

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Microaxial Flow Pump or Standard Care in Infarct-Related Cardiogenic Shock
  • Acronym: DanGer Shock
  • Year: 2024
  • Journal published in: New England Journal of Medicine
  • Citation: Møller JE, Engstrøm T, Jensen LO, Eiskjær H, Mangner N, Polzin A, et al. Microaxial Flow Pump or Standard Care in Infarct-Related Cardiogenic Shock. N Engl J Med. 2024;390:1382-1393.

Context & Rationale

  • Background
    • Infarct-related cardiogenic shock (typically in ST-segment elevation myocardial infarction) remains a high-mortality syndrome despite rapid reperfusion and modern intensive care.
    • Temporary mechanical circulatory support (tMCS) devices are widely used in practice, but prior evidence for microaxial flow pumps in acute myocardial infarction–related shock was dominated by observational studies and small trials, with persistent concerns about bleeding and vascular complications.
    • Equipoise persisted regarding whether “routine” early left ventricular unloading/support improves survival or simply shifts morbidity (e.g., haemorrhage, limb ischaemia, renal injury).
  • Research Question/Hypothesis
    • Does routine treatment with a microaxial flow pump (Impella CP) plus standard care reduce all-cause mortality at 180 days compared with standard care alone in ST-segment elevation myocardial infarction complicated by cardiogenic shock?
  • Why This Matters
    • Impella CP has been widely adopted internationally despite limited randomised evidence in infarct-related shock.
    • Defining the balance between potential survival benefit and clinically important device-related harms is essential for shock-team protocols, resource allocation, and guideline recommendations.

Design & Methods

  • Research Question: In adults with ST-segment elevation myocardial infarction and cardiogenic shock, does routine microaxial flow pump support (Impella CP) plus standard care improve 180-day survival versus standard care alone?
  • Study Type: International (Denmark and Germany), multicentre, randomised, parallel-group, open-label trial with 1:1 allocation; stratified by site; intention-to-treat primary analysis; invasive cardiac ICU/catheterisation laboratory setting.
  • Population:
    • Setting: High-volume tertiary centres with round-the-clock invasive cardiology and critical care capability.
    • Key inclusion (trial definition of infarct-related cardiogenic shock): ST-segment elevation myocardial infarction plus cardiogenic shock defined by (i) systolic blood pressure <100 mm Hg (or use of a vasopressor/inotrope) and (ii) arterial lactate ≥2.5 mmol/L and (iii) left ventricular ejection fraction <45%.
    • Key exclusions (examples): Coma after out-of-hospital cardiac arrest; overt right ventricular failure; mechanical complications of myocardial infarction; severe peripheral vascular disease precluding large-bore access; other conditions limiting feasibility or safety of device placement.
  • Intervention:
    • Device strategy: Routine microaxial flow pump (Impella CP) placement after randomisation (typically before completion of primary PCI when randomisation occurred pre-procedure).
    • Support target: Pump run at the highest possible performance level, with protocolised goal of maintaining support for at least 48 hours where feasible.
    • Care environment: Post-procedure management in specialised cardiac intensive care with protocolised haemodynamic/organ support and device weaning/removal when stabilised.
  • Comparison:
    • Standard care: Guideline-concordant shock care and revascularisation, with pharmacologic support (vasopressors/inotropes), ventilation, and escalation to other mechanical support at clinician discretion.
    • Rescue/adjunct MCS: Use of intra-aortic balloon pump and/or venoarterial extracorporeal membrane oxygenation occurred in both groups; microaxial flow pump use in the control arm was uncommon but occurred.
  • Blinding: Open-label (device strategy not amenable to participant/clinician blinding); primary endpoint (all-cause death) was objective.
  • Statistics: Sample size planned to detect an absolute mortality reduction from 60% to 42% (18% absolute) with 80% power at a two-sided 5% significance level; minimum 162 patients per group, with total target 360 to accommodate dropouts and lower-than-expected event rates; one interim analysis after 180-day follow-up of 180 patients with prespecified stopping boundary (two-sided P≤0.001). 1
  • Follow-Up Period: 180 days (primary endpoint), with additional in-hospital and 30-day outcomes reported.

Key Results

This trial was not stopped early. A prespecified interim analysis (after 180-day follow-up of the first 180 randomised patients) led the data and safety monitoring committee to recommend continuation.

Outcome Impella CP + standard care (n=179) Standard care (n=176) Effect p value / 95% CI Notes
All-cause death (primary, 180 days) 82 (45.8%) 103 (58.5%) HR 0.74 95% CI 0.55 to 0.99; P=0.04 Time-to-event analysis.
All-cause death (30 days) 69 (38.5%) 91 (51.7%) HR 0.73 95% CI 0.52 to 1.02 Secondary outcome; not statistically definitive.
Composite secondary efficacy endpoint 105 (58.7%) 122 (69.3%) HR 0.72 95% CI 0.55 to 0.93 Composite of major adverse clinical events (trial-defined).
Days alive and out of hospital (180 days) 106.6 ± 59.2 89.2 ± 70.9 Mean difference 17.4 days 95% CI 4.2 to 30.6 Higher “home time” with Impella strategy.
Composite safety endpoint 43 (24.0%) 11 (6.2%) RR 4.74 95% CI 2.36 to 9.55 Severe bleeding, limb ischaemia, haemolysis, device failure, or worsening aortic regurgitation.
Severe bleeding 36 (20.1%) 6 (3.4%) RR 6.11 95% CI 2.64 to 14.13 Defined as BARC 3 or 5 bleeding.
Renal-replacement therapy 75 (41.9%) 47 (26.7%) RR 1.98 95% CI 1.27 to 3.09 Clinically important excess organ support requirement.
Limb ischaemia requiring surgery 10 (5.6%) 2 (1.1%) RR 5.64 95% CI 1.23 to 25.81 Large-bore access complication signal.
Haemolysis 14 (7.8%) 1 (0.6%) RR 13.79 95% CI 1.81 to 104.92 Low absolute numbers; wide confidence interval.
  • Mortality was lower with the Impella CP strategy at 180 days (45.8% vs 58.5%; HR 0.74; 95% CI 0.55 to 0.99; P=0.04).
  • Device strategy increased major harms: severe bleeding (20.1% vs 3.4%) and renal-replacement therapy (41.9% vs 26.7%).
  • Protocol separation was substantial: microaxial pump placement occurred in 170/179 (95.0%) in the intervention group versus 3/176 (1.7%) crossover placement in the control group.

Internal Validity

  • Randomisation and allocation: 1:1 randomisation stratified by site; trial conduct included an independent data and safety monitoring committee; allocation concealment was operationally supported by central randomisation processes.
  • Dropout/exclusions (post-randomisation): 360 patients were randomised; 5 were excluded because informed consent could not be obtained (1 in the Impella group and 4 in the standard-care group), leaving 179 and 176 analysed.
  • Performance/detection bias: Open-label device strategy creates risk of co-intervention differences; primary outcome (all-cause death) is objective, partially mitigating detection bias.
  • Protocol adherence: Impella CP was placed in 170/179 (95.0%); 9/179 (5.0%) did not receive the device despite allocation (reasons included access/placement issues and evolving clinical judgement).
  • Crossover/contamination: In the standard-care group, 3/176 (1.7%) received Impella CP as crossover treatment; an additional 4 patients received Impella CP as an unloading strategy during venoarterial extracorporeal membrane oxygenation.
  • Baseline comparability (selected key markers): Age 67 vs 68 years; arterial lactate 3.3 vs 3.4 mmol/L; systolic blood pressure 80 vs 80 mm Hg; left ventricular ejection fraction 29.0% vs 30.0%.
  • Heterogeneity: Prespecified subgroup analyses did not identify clear effect modification across major clinical strata; the trial was conducted in two high-resource health systems with specialised shock pathways.
  • Timing: Randomisation occurred before revascularisation in 74/179 (41.3%) vs 85/176 (48.3%); when randomisation occurred before revascularisation, Impella CP placement was achieved before the procedure in 66/74 (89.2%) and before PCI completion in 84/89 (94.4%).
  • Dose/intensity of intervention: Median duration of Impella CP support was 59 hours (IQR 30 to 87); device failure occurred in 4/179 (2.2%).
  • Separation of the variable of interest: Microaxial pump placement 170/179 (95.0%) vs 3/176 (1.7%); venoarterial ECMO 21/179 (11.7%) vs 33/176 (18.8%); intra-aortic balloon pump 5/179 (2.8%) vs 5/176 (2.8%).
  • Outcome assessment: Primary endpoint was all-cause mortality at 180 days; the objectivity of mortality strengthens outcome ascertainment.
  • Statistical rigor: Time-to-event methods used for mortality; prespecified interim analysis and alpha spending were incorporated (primary outcome P value met the reported threshold).

Conclusion on Internal Validity: Overall, internal validity appears moderate-to-strong given robust randomisation, minimal crossover, and an objective primary outcome; limitations include open-label co-intervention potential and small post-randomisation exclusions related to consent.

External Validity

  • Population representativeness: Highly selected infarct-related cardiogenic shock population in expert centres; among 1211 screened, 851 were excluded, with coma after out-of-hospital cardiac arrest a common exclusion (435 patients).
  • Exclusions with major generalisability impact: Exclusion of comatose post–cardiac arrest patients limits applicability to many real-world shock presentations where hypoxic brain injury drives outcomes.
  • System/setting dependence: Implementation requires immediate access to large-bore arterial cannulation expertise, catheter laboratory workflows for rapid device placement, and capability to manage bleeding/vascular complications and renal failure.
  • Applicability to other shock phenotypes: Findings apply most directly to ST-segment elevation myocardial infarction–related shock with predominant left ventricular failure; extrapolation to non–ST-segment elevation myocardial infarction shock, predominant right ventricular failure, or mixed aetiologies is uncertain.

Conclusion on External Validity: Generalisability is moderate-to-limited: the results are most applicable to carefully selected infarct-related cardiogenic shock patients managed in experienced, high-resource centres with established shock pathways.

Strengths & Limitations

  • Strengths: Large pragmatic randomised trial in a high-mortality condition; clinically meaningful follow-up (180 days); objective primary endpoint; strong protocol separation (95.0% vs 1.7% microaxial pump placement); comprehensive safety reporting.
  • Limitations: Open-label design with potential co-intervention imbalance; substantial exclusions during screening (notably coma after out-of-hospital cardiac arrest); post-randomisation exclusions due to consent issues; high rate of major device-related complications; recruitment over many years with evolving practice patterns.

Interpretation & Why It Matters

  • Clinical meaning
    Routine Impella CP support in selected infarct-related cardiogenic shock was associated with improved survival at 180 days, but with a substantial increase in major harms (notably severe bleeding and renal-replacement therapy), shifting practice from “device enthusiasm” to “device plus selection plus systems”.
  • Mechanistic inference
    The survival signal is compatible with haemodynamic stabilisation and left ventricular unloading facilitating myocardial recovery and end-organ perfusion, yet the competing morbidity profile highlights that procedural and device-related complications can erode net benefit in less selected populations or less experienced centres.

Controversies & Subsequent Evidence

  • Benefit–harm tension as the central controversy: The trial demonstrated lower 180-day mortality with Impella CP but a markedly higher composite safety endpoint (24.0% vs 6.2%) and severe bleeding (20.1% vs 3.4%), reframing “routine use” as a strategy that may only be acceptable within tightly managed shock programmes.
  • Selection and generalisability: Exclusion of comatose post–out-of-hospital cardiac arrest patients (a large fraction of screened patients) creates a selected population in whom survival benefit may be more achievable; this aligns with subsequent synthesis suggesting treatment effects for tMCS concentrate in infarct-related shock without hypoxic brain injury. 5
  • Timing of support relative to PCI: A substantial proportion were randomised after revascularisation (58.7% vs 51.7%), raising questions about whether observed benefit reflects post-PCI support rather than “pre-PCI unloading” paradigms; subgroup analyses did not show a clear interaction by timing.
  • Correspondence-driven critique: Published letters questioned net clinical benefit given complication rates and the representativeness of the enrolled population; the authors’ reply emphasised prespecified endpoints, systems-level expertise, and the necessity of weighing harms against mortality benefit. 3
  • Editorial framing: The accompanying editorial highlighted the trial as a major advance for an evidence-scarce domain, while stressing that outcomes likely depend on patient selection, procedural expertise, and integrated shock-team care rather than the device alone. 2
  • Comparative MCS evidence: In infarct-related cardiogenic shock, early venoarterial ECMO has not demonstrated a survival advantage and increases complications, reinforcing that escalation to more invasive support is not automatically beneficial and that modality/selection matter. 4
  • Meta-analytic synthesis incorporating DanGer Shock: A recent systematic review/meta-analysis that included DanGer Shock reported mortality reduction signals for microaxial pump strategies alongside increased bleeding/vascular complications, consistent with the trial’s benefit–harm profile. 6
  • Network-level synthesis: Contemporary comparative analyses (including network meta-analysis approaches combining randomised and propensity-matched evidence) underline uncertainty and heterogeneity in device effects, and emphasise complications as a dominant competing risk. 7
  • Guideline traction: Post-trial guideline updates for acute coronary syndromes incorporate evolving evidence around tMCS and emphasise shock-system readiness and careful patient selection when considering microaxial pumps. 8

Summary

  • DanGer Shock randomised 360 patients with infarct-related cardiogenic shock; 355 were analysed (179 vs 176).
  • Routine Impella CP support reduced all-cause mortality at 180 days (45.8% vs 58.5%; HR 0.74; 95% CI 0.55 to 0.99; P=0.04).
  • Major harms were increased with Impella CP: severe bleeding (20.1% vs 3.4%) and renal-replacement therapy (41.9% vs 26.7%).
  • Protocol separation was large (Impella CP placed in 95.0% vs 1.7% crossover), supporting interpretability of the device-strategy comparison.
  • Generalisability is constrained by exclusion of comatose post–out-of-hospital cardiac arrest patients and the requirement for experienced high-resource shock systems.

Further Reading

Other Trials

Systematic Review & Meta Analysis

Observational Studies

Guidelines

Notes

  • Only references with verifiable DOI links from the available source set are listed above; additional relevant observational cohorts and position papers exist but their DOI details were not available in the provided materials.

Overall Takeaway

DanGer Shock provides rare randomised evidence that a routine Impella CP strategy can improve 180-day survival in carefully selected infarct-related cardiogenic shock patients managed in expert centres. The price is a substantial increase in major complications (especially severe bleeding and renal-replacement therapy), making patient selection, procedural expertise, and system readiness central to translating the mortality benefit into real-world net clinical benefit.

Overall Summary

  • Impella CP strategy reduced 180-day mortality (45.8% vs 58.5%), but increased severe bleeding (20.1% vs 3.4%) and renal-replacement therapy (41.9% vs 26.7%).
  • High protocol separation and objective mortality endpoint strengthen causal inference; open-label design and selective enrolment constrain generalisability.
  • Best interpreted as evidence for “device + selection + shock systems”, not indiscriminate routine use.

Bibliography